Underground pipe network monitoring method and system

Abstract

The invention discloses an underground pipe network monitoring method and system, and relates to the technical field of underground pipe network monitoring, and the method comprises the steps: arranging pressure sensors and flowmeters at key nodes of an underground pipe network, and collecting the pressure time sequence data and flow time sequence data of each node; calculating the hydraulic impedance of each pipe section according to the pressure time sequence data and the flow time sequence data, and constructing a hydraulic impedance space-time field covering a pipe network topology and a time dimension; extracting a space change trend and a time change trend based on the hydraulic impedance space-time field, fusing to generate a space-time gradient anomaly index, identifying a primary leakage suspicious pipe section according to the index, and forming a primary leakage alarm event; the method comprises the following steps: determining water hammer wave propagation characteristics in combination with a topological structure of an underground pipe network and pipe section physical attributes, performing water hammer wave propagation rule consistency verification on a plurality of alarm events by using time and position information of a primary leakage alarm event, and screening out an effective leakage event cluster conforming to physical propagation logic.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of underground pipe network monitoring, in particular to an underground pipe network monitoring method and system. BACKGROUND

[0002] The underground pipe network monitoring technology refers to a comprehensive technical system using sensors, communication, data analysis and modeling to realize real-time sensing, abnormal identification, fault positioning and risk assessment of the operation state of underground pipe systems such as water supply, drainage, gas and heat supply. Therefore, how to use advanced technical means to improve the intelligent level and safety of underground pipe network monitoring has become one of the problems to be solved at present.

[0003] The traditional method mainly depends on single parameters such as pressure drop and flow anomaly for judgment, is easily disturbed by normal working conditions such as water fluctuation and water pump start-stop, is difficult to distinguish between real leakage and non-leakage disturbance, and the existing alarm is often triggered based on local threshold, without considering the hydraulic propagation law, resulting in frequent isolated and contradictory alarms, which cannot form credible event correlation. At the same time, the signal change caused by small leakage is weak and often submerged by noise. The traditional method lacks the modeling ability of spatio-temporal coupling characteristics and is difficult to capture the slow development or localized abnormal trend. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides an underground pipe network monitoring method to solve the problem that the existing alarm is often triggered based on local threshold, without considering the hydraulic propagation law, resulting in frequent isolated and contradictory alarms, which cannot form credible event correlation. At the same time, the signal change caused by small leakage is weak and often submerged by noise. The traditional method lacks the modeling ability of spatio-temporal coupling characteristics and is difficult to capture the slow development or localized abnormal trend.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides an underground pipe network monitoring method, which comprises:

[0008] Pressure sensors and flow meters are arranged at key nodes of the underground pipe network to collect pressure time series data and flow time series data of each node;

[0009] The hydraulic impedance of each pipe section is calculated according to the pressure time series data and the flow time series data, and a hydraulic impedance spatio-temporal field covering the topology of the pipe network and the time dimension is constructed;

[0010] The spatial variation trend and the temporal variation trend are extracted based on the hydraulic impedance spatio-temporal field, a spatio-temporal gradient anomaly index is fused and generated, and a preliminary leakage suspicious pipe section is identified according to the index to form a primary leakage alarm event.

[0011] The propagation characteristics of water hammer waves are determined by combining the topology of the underground pipe network and the physical properties of the pipe segments. The time and location information of the primary leakage alarm events are used to verify the consistency of the water hammer wave propagation law of multiple alarm events, and the effective leakage event clusters that conform to the physical propagation logic are selected.

[0012] The pressure time series data, flow time series data, and the effective leakage event cluster are used as boundary conditions to drive the lightweight digital twin model to invert the pipeline network state parameters in areas where no sensors are deployed.

[0013] Based on the spatiotemporal gradient anomaly index, the attribution information of the effective leakage event cluster, and the pipeline status parameters obtained by the lightweight digital twin model inversion, the leakage probability of each pipeline segment is comprehensively evaluated, and the final leakage diagnosis result is output.

[0014] As a preferred embodiment of the underground pipe network monitoring method of the present invention, the specific steps of calculating the hydraulic impedance of each pipe segment based on pressure time series data and flow time series data, and constructing a hydraulic impedance spatiotemporal field covering the pipe network topology and time dimension are as follows:

[0015] For any pipe segment with upstream and downstream pressure measurement points, the upstream node pressure value, the downstream node pressure value, and the flow rate through the pipe segment are obtained at the sampling time.

[0016] The pressure difference is obtained by subtracting the pressure value of the downstream node from the pressure value of the upstream node. Then, the pressure difference is divided by the flow rate at the corresponding moment to obtain the hydraulic impedance of the pipe segment at that moment. The expression is as follows:

[0017] ;

[0018] in, Indicates pipe section At sampling time hydraulic resistance, Indicates pipe section Upstream node at time Pressure value, Indicates pipe section Downstream nodes at time Pressure value, Indicates pipe section At any moment The flow rate value;

[0019] Repeat the above process, traversing all pipe sections with complete measurement conditions and all sampling times, to form a two-dimensional data structure consisting of pipe section location, time index, and hydraulic impedance value, which serves as the spatiotemporal field of hydraulic impedance.

[0020] As a preferred embodiment of the underground pipe network monitoring method of the present invention, the steps of extracting spatial and temporal variation trends based on the spatiotemporal field of hydraulic impedance, fusing them to generate a spatiotemporal gradient anomaly index, and identifying preliminary suspected leaking pipe sections based on this index to form a primary leak alarm event are as follows:

[0021] At a fixed sampling time, differential calculations are performed on the hydraulic impedance values ​​of adjacent pipe sections along the pipe network topology to obtain the spatial variation trend.

[0022] Under a fixed pipe section, differential calculations are performed on the hydraulic impedance values ​​at continuous sampling times to obtain the time variation trend;

[0023] The spatial and temporal variation trends are weighted according to a preset ratio to synthesize a single numerical index, which serves as the spatiotemporal gradient anomaly index. The expression is as follows:

[0024] ;

[0025] in, Indicates pipe section At any moment The spatiotemporal gradient anomaly index, This indicates the spatial variation trend, reflecting the sudden change in hydraulic impedance of adjacent pipe sections at the same time. This indicates the trend over time, reflecting the rate of change of hydraulic impedance of the same pipe section at adjacent moments. Preset weighting coefficients;

[0026] When the index exceeds the first threshold determined based on historical normal operating conditions, the corresponding pipe section is marked as a preliminary suspected leak pipe section, and its location and trigger time are recorded to form a primary leak alarm event.

[0027] As a preferred embodiment of the underground pipeline network monitoring method of the present invention, the following steps are taken: Determining the water hammer wave propagation characteristics by combining the topology and physical properties of the underground pipeline network, utilizing the time and location information of primary leakage alarm events, performing consistency verification of water hammer wave propagation patterns on multiple alarm events, and filtering out valid leakage event clusters that conform to physical propagation logic:

[0028] Based on the material, diameter, and wall thickness parameters of each pipe section, determine the propagation speed of the water hammer wave in that pipe section;

[0029] Based on the pipeline topology, calculate the shortest path between any two primary leak alarm events and the corresponding pipe segments. This path consists of a series of continuous pipe segments.

[0030] Divide the length of each pipe segment along the path by its corresponding water hammer wave propagation velocity and sum them up to obtain the theoretical water hammer wave propagation delay, expressed as follows:

[0031] ;

[0032] in, Indicates a primary leak alert event The incident was propagated to the corresponding pipe section. The theoretical water hammer wave propagation delay for the corresponding pipe section, This indicates the number of pipe segments included in the shortest path. Indicates the first in the path The physical length of the pipe segment Indicates the water hammer wave at the 1st The propagation velocity within a pipe segment is determined by the material, diameter, and wall thickness of that segment.

[0033] Compare the actual time difference between any two primary leakage alarm events with the theoretical propagation delay. If the absolute value of the difference does not exceed the preset time tolerance, then the event is determined to satisfy the physical propagation law.

[0034] All primary leak alarm events that mutually satisfy propagation constraints are aggregated into the same valid leak event cluster.

[0035] As a preferred embodiment of the underground pipeline monitoring method of the present invention, the step of using pressure time-series data, flow time-series data, and the effective leakage event cluster as boundary conditions to drive the lightweight digital twin model to invert the pipeline state parameters in areas where no sensors are deployed is as follows:

[0036] Construct a simplified hydraulic simulation model that is consistent with the actual underground pipe network topology;

[0037] Use the node pressure time series data and flow time series data with installed sensors as the model input boundary;

[0038] The pipe segment identifiers and occurrence times included in the effective leakage event clusters are used as local disturbance sources and introduced into the flow imbalance terms at the corresponding locations in the model.

[0039] The model's output pressure at known nodes is solved iteratively to ensure that the error between the measured pressure and the output pressure is less than the convergence tolerance.

[0040] After the model converges, the pressure, flow rate and hydraulic impedance values ​​of the pipe section without installed sensors are extracted as inversion state parameters.

[0041] The standardized deviation of the inverted hydraulic impedance from the normal reference is calculated using the following formula:

[0042] ;

[0043] in, Indicates pipe section The standardized deviation, This represents the pipe segment obtained from the inversion of the lightweight digital twin model. The average hydraulic impedance, Indicates pipe section The average hydraulic impedance under normal historical operating conditions It represents the standard deviation of hydraulic impedance under normal historical operating conditions.

[0044] As a preferred embodiment of the underground pipeline monitoring method of the present invention, the steps of comprehensively assessing the leakage probability of each pipe section based on the spatiotemporal gradient anomaly index, the attribution information of the effective leakage event cluster, and the pipeline state parameters obtained by inversion from the lightweight digital twin model, and outputting the final leakage diagnosis result, are as follows:

[0045] For each pipe segment, the normalized maximum spatiotemporal gradient anomaly index, the effective leakage event cluster attribution indicator, and the standardized deviation of the inverted hydraulic impedance are calculated respectively.

[0046] The three indicators are linearly weighted according to preset weights to obtain a comprehensive leakage probability score, expressed as:

[0047] ;

[0048] in, Indicates pipe section The comprehensive leakage probability score, Represents the normalized maximum spatiotemporal gradient anomaly index. For effective leak event cluster attribution indication, if the pipe section Belongs to any valid leakage event cluster, , , These are non-negative weighting coefficients;

[0049] When the comprehensive leakage probability score exceeds the second threshold, it is determined that there is a leak in the pipe section;

[0050] The output includes pipe segment identification, comprehensive leakage probability score, recommended inspection priority, and structured diagnostic results of the suspected leak initiation time.

[0051] As a preferred embodiment of the underground pipe network monitoring method of the present invention, wherein: the average hydraulic impedance under historical normal operating conditions The standard deviation, the standard deviation of the spatiotemporal gradient anomaly index, the first threshold and the second threshold are all dynamically updated based on the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

[0052] Secondly, the present invention provides an underground pipeline network monitoring system, comprising:

[0053] The module includes a hydraulic impedance field construction module, a spatiotemporal gradient anomaly detection module, a water hammer wave event verification module, a digital twin state inversion module, a comprehensive leakage assessment module, and an adaptive threshold update module.

[0054] The hydraulic impedance field construction module is used to calculate the hydraulic impedance of each pipe section based on pressure time series data and flow time series data, and to construct a hydraulic impedance spatiotemporal field covering the pipe network topology and time dimension.

[0055] The spatiotemporal gradient anomaly detection module is used to extract spatial and temporal variation trends based on the spatiotemporal field of hydraulic impedance, fuse them to generate a spatiotemporal gradient anomaly index, and identify suspected leaking pipe sections based on the index to form a primary leak alarm event.

[0056] The water hammer wave event verification module is used to determine the water hammer wave propagation characteristics by combining the topology of the underground pipe network and the physical properties of the pipe segments, to verify the consistency of the water hammer wave propagation law of multiple primary leakage alarm events, and to screen out the effective leakage event clusters that conform to the physical propagation logic.

[0057] The digital twin state inversion module is used to drive the lightweight digital twin model to run by using pressure time series data, flow time series data and effective leakage event clusters as boundary conditions, and invert the pipeline network state parameters in areas where no sensors are deployed.

[0058] The comprehensive leakage assessment module is used to calculate the comprehensive leakage probability score of each pipe segment based on the spatiotemporal gradient anomaly index, the attribution information of effective leakage event clusters, and the pipeline status parameters obtained by inversion from the digital twin model, and output the final leakage diagnosis result.

[0059] The adaptive threshold update module is used to dynamically update the mean and standard deviation of hydraulic impedance, the standard deviation of spatiotemporal gradient anomaly index, the first threshold and the second threshold under historical normal operating conditions based on the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

[0060] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein the computer program, when executed by the processor, implements any step of the underground pipeline monitoring method as described in the first aspect of the present invention.

[0061] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the underground pipeline monitoring method as described in the first aspect of the present invention.

[0062] The beneficial effects of this invention are as follows: by constructing a hydraulic impedance spatiotemporal field and introducing a spatiotemporal gradient anomaly index, the early sensitivity to minor leaks is significantly improved. By combining the physical propagation law of water hammer waves to perform consistency verification on multi-point alarm events, false alarms caused by transient disturbances such as pump start-up and shutdown and valve operation are effectively suppressed, and the positioning accuracy is greatly improved. By integrating a lightweight digital twin model, the system can achieve high-precision inversion of the status of the pipeline network in unmonitored areas under the condition of sparse sensor deployment, breaking through the practical bottleneck of missing old pipeline network maps and large sensing blind spots. At the same time, through multi-source evidence weighted fusion and threshold adaptive mechanism, the system maintains high robustness and diagnostic reliability under different seasons, loads and pipe ages. Attached Figure Description

[0063] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a flowchart of the underground pipeline monitoring method in Example 1.

[0065] Figure 2 This is a schematic diagram of the underground pipeline monitoring system in Example 1. Detailed Implementation

[0066] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0067] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0068] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0069] Example 1, referring to Figure 1 and Figure 2 This embodiment of the invention provides a method for monitoring underground pipe networks, comprising the following steps:

[0070] S1. Install pressure sensors and flow meters at key nodes of the underground pipe network to collect pressure time-series data and flow time-series data at each node;

[0071] It should be noted that by deploying pressure sensors and flow meters at key nodes, high-precision pressure and flow time-series data can be acquired in real time and synchronously, providing a reliable data foundation for subsequent hydraulic state modeling and anomaly identification, and effectively avoiding misjudgments caused by missing or delayed data.

[0072] S2. Calculate the hydraulic impedance of each pipe section based on pressure time series data and flow time series data, and construct a spatiotemporal field of hydraulic impedance covering the pipe network topology and time dimension.

[0073] Furthermore, for any pipe segment with upstream and downstream pressure measurement points, the upstream node pressure value, the downstream node pressure value, and the flow rate through the pipe segment are obtained at the sampling time.

[0074] The pressure difference is obtained by subtracting the pressure value of the downstream node from the pressure value of the upstream node. Then, the pressure difference is divided by the flow rate at the corresponding moment to obtain the hydraulic impedance of the pipe segment at that moment. The expression is as follows:

[0075] ;

[0076] in, Indicates pipe section At sampling time hydraulic resistance, Indicates pipe section Upstream node at time Pressure value, Indicates pipe section Downstream nodes at time Pressure value, Indicates pipe section At any moment The flow rate value;

[0077] Repeat the above process to traverse all pipe sections with complete measurement conditions and all sampling times, forming a two-dimensional data structure consisting of pipe section location, time index, and hydraulic impedance value, which serves as the spatiotemporal field of hydraulic impedance.

[0078] It should be noted that by dynamically calculating hydraulic impedance based on the ratio of upstream and downstream pressure difference to flow rate, and constructing a spatiotemporal field of hydraulic impedance covering the pipeline network topology and time dimension, the complex pipeline network operation status can be transformed into a structured, quantifiable, and comparable unified indicator system, thereby improving the sensitivity and characterization ability of local anomalies (such as leakage and blockage).

[0079] S3. Based on the spatiotemporal field of hydraulic impedance, extract the spatial and temporal variation trends, fuse them to generate a spatiotemporal gradient anomaly index, and identify the pipe sections suspected of initial leakage based on the index, forming a primary leakage alarm event.

[0080] Furthermore, at a fixed sampling time, differential calculations are performed on the hydraulic impedance values ​​of adjacent pipe sections along the pipeline network topology to obtain the spatial variation trend;

[0081] Under a fixed pipe section, differential calculations are performed on the hydraulic impedance values ​​at continuous sampling times to obtain the time variation trend;

[0082] The spatial and temporal variation trends are weighted according to a preset ratio to synthesize a single numerical index, which serves as the spatiotemporal gradient anomaly index. The expression is as follows:

[0083] ;

[0084] in, Indicates pipe section At any moment The spatiotemporal gradient anomaly index, This indicates the spatial variation trend, reflecting the sudden change in hydraulic impedance of adjacent pipe sections at the same time. This indicates the trend over time, reflecting the rate of change of hydraulic impedance of the same pipe section at adjacent moments. Preset weighting coefficients;

[0085] When the index exceeds the first threshold determined based on historical normal operating conditions, the corresponding pipe section is marked as a preliminary suspected leak pipe section, and its location and trigger time are recorded to form a primary leak alarm event.

[0086] It should be noted that by jointly analyzing the spatial abruptness and temporal rate of change of hydraulic impedance and integrating them into a spatiotemporal gradient anomaly index, it is possible to effectively distinguish between abnormal patterns caused by actual leaks and disturbances caused by normal operating conditions such as water usage fluctuations. This allows for the accurate screening of physically significant suspicious pipe sections in the early stages, thereby reducing the false alarm rate.

[0087] S4. Combine the topology of the underground pipe network and the physical properties of the pipe sections to determine the propagation characteristics of water hammer waves. Use the time and location information of the primary leakage alarm events to verify the consistency of the water hammer wave propagation law of multiple alarm events and screen out the effective leakage event clusters that conform to the physical propagation logic.

[0088] Furthermore, based on the material, diameter, and wall thickness parameters of each pipe section, the propagation speed of the water hammer wave in that pipe section is determined.

[0089] Based on the pipeline topology, calculate the shortest path between any two primary leak alarm events and the corresponding pipe segments. This path consists of a series of continuous pipe segments.

[0090] Divide the length of each pipe segment along the path by its corresponding water hammer wave propagation velocity and sum them up to obtain the theoretical water hammer wave propagation delay, expressed as follows:

[0091] ;

[0092] in, Indicates a primary leak alert event The incident was propagated to the corresponding pipe section. The theoretical water hammer wave propagation delay for the corresponding pipe section, This indicates the number of pipe segments included in the shortest path. Indicates the first in the path The physical length of the pipe segment Indicates the water hammer wave at the 1st The propagation velocity within a pipe segment is determined by the material, diameter, and wall thickness of that segment.

[0093] Compare the actual time difference between any two primary leakage alarm events with the theoretical propagation delay. If the absolute value of the difference does not exceed the preset time tolerance, then the event is determined to satisfy the physical propagation law.

[0094] Aggregate all primary leakage alarm events that mutually satisfy propagation constraints into a single valid leakage event cluster;

[0095] It should be noted that by using the physical properties of the pipeline network to determine the propagation speed of water hammer waves and performing a propagation delay consistency check on primary alarm events, isolated alarms caused by noise, instrument errors, or non-leakage disturbances can be filtered out from the physical mechanism level, retaining only event clusters that conform to the laws of hydraulic propagation, thereby improving the reliability and confidence of leak event determination.

[0096] S5. Using pressure time series data, flow time series data, and the effective leakage event cluster as boundary conditions, drive the lightweight digital twin model to run and invert the pipeline network state parameters in areas where no sensors are deployed.

[0097] Furthermore, a simplified hydraulic simulation model consistent with the actual underground pipe network topology is constructed;

[0098] Use the node pressure time series data and flow time series data with installed sensors as the model input boundary;

[0099] The pipe segment identifiers and occurrence times included in the effective leakage event clusters are used as local disturbance sources and introduced into the flow imbalance terms at the corresponding locations in the model.

[0100] The model's output pressure at known nodes is solved iteratively to ensure that the error between the measured pressure and the output pressure is less than the convergence tolerance.

[0101] After the model converges, the pressure, flow rate and hydraulic impedance values ​​of the pipe section without installed sensors are extracted as inversion state parameters.

[0102] The standardized deviation of the inverted hydraulic impedance from the normal reference is calculated using the following formula:

[0103] ;

[0104] in, Indicates pipe section The standardized deviation, This represents the pipe segment obtained from the inversion of the lightweight digital twin model. The average hydraulic impedance, Indicates pipe section The average hydraulic impedance under normal historical operating conditions This represents the standard deviation of hydraulic impedance under historical normal operating conditions.

[0105] It should be noted that by using measured data and valid leakage event clusters as boundary conditions to drive the lightweight digital twin model, high-fidelity state inversion can be achieved in areas without sensor coverage. This not only expands the perception capability of monitoring blind spots, but also provides quantifiable leakage risk data for untested pipe sections, effectively improving the completeness and practicality of network-wide diagnosis.

[0106] S6. Based on the spatiotemporal gradient anomaly index, the attribution information of the effective leakage event cluster, and the pipeline status parameters obtained by the lightweight digital twin model inversion, comprehensively evaluate the leakage probability of each pipeline segment and output the final leakage diagnosis result.

[0107] Furthermore, for each pipe segment, the normalized maximum spatiotemporal gradient anomaly index, the effective leakage event cluster attribution indicator, and the standardized deviation of the inverted hydraulic impedance are calculated separately.

[0108] The three indicators are linearly weighted according to preset weights to obtain a comprehensive leakage probability score, expressed as:

[0109] ;

[0110] in, Indicates pipe section The comprehensive leakage probability score, Represents the normalized maximum spatiotemporal gradient anomaly index. For effective leak event cluster attribution indication, if the pipe section Belongs to any valid leakage event cluster, , , These are non-negative weighting coefficients;

[0111] When the comprehensive leakage probability score exceeds the second threshold, it is determined that there is a leak in the pipe section;

[0112] The output includes pipe segment identification, comprehensive leakage probability score, recommended inspection priority, and structured diagnostic results of the suspected leak initiation time;

[0113] Average hydraulic impedance under normal historical operating conditions The standard deviation, the standard deviation of the spatiotemporal gradient anomaly index, the first threshold and the second threshold are all dynamically updated according to the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

[0114] It should be noted that by integrating the spatiotemporal anomaly intensity, event cluster attribution logic, and inversion state deviation as triple evidence, and combining them with a dynamic threshold mechanism for weighted evaluation, a comprehensive leakage probability score with strong interpretability and high robustness can be generated. This ensures that the final diagnostic results reflect both the current degree of anomaly and the adaptability to historical operating conditions, providing scientific and operable priority guidance for operation and maintenance decisions.

[0115] This embodiment also provides an underground pipeline network monitoring system, including:

[0116] The module includes a hydraulic impedance field construction module, a spatiotemporal gradient anomaly detection module, a water hammer wave event verification module, a digital twin state inversion module, a comprehensive leakage assessment module, and an adaptive threshold update module.

[0117] The hydraulic impedance field construction module is used to calculate the hydraulic impedance of each pipe section based on pressure time series data and flow time series data, and to construct a hydraulic impedance spatiotemporal field covering the pipe network topology and time dimension.

[0118] The spatiotemporal gradient anomaly detection module is used to extract spatial and temporal variation trends based on the spatiotemporal field of hydraulic impedance, fuse them to generate a spatiotemporal gradient anomaly index, and identify suspected pipe sections with initial leakage based on the index to form a primary leakage alarm event.

[0119] The water hammer wave event verification module is used to determine the propagation characteristics of water hammer waves by combining the topology of the underground pipe network and the physical properties of the pipe segments. It performs consistency verification of the water hammer wave propagation law for multiple primary leakage alarm events and filters out valid leakage event clusters that conform to the physical propagation logic.

[0120] The digital twin state inversion module is used to drive the lightweight digital twin model to run by using pressure time series data, flow time series data and effective leakage event clusters as boundary conditions, and invert the pipeline network state parameters in areas where no sensors are deployed.

[0121] The comprehensive leakage assessment module is used to calculate the comprehensive leakage probability score of each pipe segment based on the spatiotemporal gradient anomaly index, the attribution information of effective leakage event clusters, and the pipeline status parameters obtained by inversion from the digital twin model, and output the final leakage diagnosis result.

[0122] The adaptive threshold update module is used to dynamically update the mean and standard deviation of hydraulic impedance, the standard deviation of spatiotemporal gradient anomaly index, the first threshold and the second threshold under historical normal operating conditions based on the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

[0123] In summary, this invention significantly improves the early sensitivity to minor leaks by constructing a hydraulic impedance spatiotemporal field and introducing a spatiotemporal gradient anomaly index. By combining the physical propagation law of water hammer waves to perform consistency verification on multi-point alarm events, it effectively suppresses false alarms caused by transient disturbances such as pump start-up and shutdown and valve operation, greatly improving positioning accuracy. By integrating a lightweight digital twin model, it achieves high-precision inversion of the status of pipeline networks in unmonitored areas under sparse sensor deployment conditions, breaking through the practical bottlenecks of missing old pipeline network maps and large sensing blind spots. At the same time, through multi-source evidence weighted fusion and threshold adaptive mechanism, the system maintains high robustness and diagnostic reliability under different seasons, loads, and pipe age conditions.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for monitoring underground pipe networks, characterized in that: include: Pressure sensors and flow meters are installed at key nodes of the underground pipe network to collect pressure time-series data and flow time-series data at each node. The hydraulic impedance of each pipe section is calculated based on pressure time series data and flow time series data, and a spatiotemporal field of hydraulic impedance covering the pipe network topology and time dimension is constructed. Based on the extraction of spatial and temporal variation trends from the spatiotemporal field of hydraulic impedance, a spatiotemporal gradient anomaly index is generated and the pipe section suspected of initial leakage is identified based on the index, thus forming a primary leakage alarm event. The propagation characteristics of water hammer waves are determined by combining the topology of the underground pipe network and the physical properties of the pipe segments. The time and location information of the primary leakage alarm events are used to verify the consistency of the water hammer wave propagation law of multiple alarm events, and the effective leakage event clusters that conform to the physical propagation logic are selected. The pressure time series data, flow time series data, and the effective leakage event cluster are used as boundary conditions to drive the lightweight digital twin model to invert the pipeline network state parameters in areas where no sensors are deployed. Based on the spatiotemporal gradient anomaly index, the attribution information of the effective leakage event cluster, and the pipeline status parameters obtained by the lightweight digital twin model inversion, the leakage probability of each pipeline segment is comprehensively evaluated, and the final leakage diagnosis result is output.

2. The underground pipeline monitoring method as described in claim 1, characterized in that: The steps for calculating the hydraulic impedance of each pipe segment based on pressure and flow time series data, and constructing a spatiotemporal field of hydraulic impedance covering the pipe network topology and time dimension, are as follows: For any pipe segment with upstream and downstream pressure measurement points, the upstream node pressure value, the downstream node pressure value, and the flow rate through the pipe segment are obtained at the sampling time. The pressure difference is obtained by subtracting the pressure value of the downstream node from the pressure value of the upstream node. Then, the pressure difference is divided by the flow rate at the corresponding moment to obtain the hydraulic impedance of the pipe segment at that moment. The expression is as follows: ； in, Indicates pipe section At sampling time hydraulic resistance, Indicates pipe section Upstream node at time Pressure value, Indicates pipe section Downstream nodes at time Pressure value, Indicates pipe section At any moment The flow rate value; Repeat the above process, traversing all pipe sections with complete measurement conditions and all sampling times, to form a two-dimensional data structure consisting of pipe section location, time index, and hydraulic impedance value, which serves as the spatiotemporal field of hydraulic impedance.

3. The underground pipeline monitoring method as described in claim 2, characterized in that: The process involves extracting spatial and temporal variation trends based on the spatiotemporal field of hydraulic impedance, fusing them to generate a spatiotemporal gradient anomaly index, and identifying initially suspected leaking pipe sections based on this index, thus forming a primary leak alarm event. The specific steps are as follows: At a fixed sampling time, differential calculations are performed on the hydraulic impedance values ​​of adjacent pipe sections along the pipe network topology to obtain the spatial variation trend. Under a fixed pipe section, differential calculations are performed on the hydraulic impedance values ​​at continuous sampling times to obtain the time variation trend; The spatial and temporal variation trends are weighted according to a preset ratio to synthesize a single numerical index, which serves as the spatiotemporal gradient anomaly index. The expression is as follows: ; in, Indicates pipe section At any moment The spatiotemporal gradient anomaly index, This indicates the spatial variation trend, reflecting the sudden change in hydraulic impedance of adjacent pipe sections at the same time. This indicates the trend over time, reflecting the rate of change of hydraulic impedance of the same pipe section at adjacent moments. Preset weighting coefficients; When the index exceeds the first threshold determined based on historical normal operating conditions, the corresponding pipe section is marked as a preliminary suspected leak pipe section, and its location and trigger time are recorded to form a primary leak alarm event.

4. The underground pipeline monitoring method as described in claim 3, characterized in that: The process involves determining the water hammer wave propagation characteristics by combining the topology of the underground pipe network and the physical properties of the pipe segments. Using the time and location information of primary leak alarm events, a consistency check of the water hammer wave propagation patterns is performed on multiple alarm events to filter out valid leak event clusters that conform to the physical propagation logic. The specific steps are as follows: Based on the material, diameter, and wall thickness parameters of each pipe section, determine the propagation speed of the water hammer wave in that pipe section; Based on the pipeline topology, calculate the shortest path between any two primary leak alarm events and the corresponding pipe segments. This path consists of a series of continuous pipe segments. Divide the length of each pipe segment along the path by its corresponding water hammer wave propagation velocity and sum them up to obtain the theoretical water hammer wave propagation delay, expressed as follows: ; in, Indicates a primary leak alert event The incident was propagated to the corresponding pipe section. The theoretical water hammer wave propagation delay for the corresponding pipe section, This indicates the number of pipe segments included in the shortest path. Indicates the first in the path The physical length of the pipe segment Indicates the water hammer wave at the 1st The propagation velocity within a pipe segment is determined by the material, diameter, and wall thickness of that segment. Compare the actual time difference between any two primary leakage alarm events with the theoretical propagation delay. If the absolute value of the difference does not exceed the preset time tolerance, then the event is determined to satisfy the physical propagation law. All primary leak alarm events that mutually satisfy propagation constraints are aggregated into the same valid leak event cluster.

5. The underground pipeline monitoring method as described in claim 4, characterized in that: The step of using pressure time-series data, flow time-series data, and the effective leakage event cluster as boundary conditions to drive the lightweight digital twin model to invert the pipeline network state parameters in areas without sensor deployment is as follows: Construct a simplified hydraulic simulation model that is consistent with the actual underground pipe network topology; Use the node pressure time series data and flow time series data with installed sensors as the model input boundary; The pipe segment identifiers and occurrence times included in the effective leakage event clusters are used as local disturbance sources and introduced into the flow imbalance terms at the corresponding locations in the model. The model's output pressure at known nodes is solved iteratively to ensure that the error between the measured pressure and the output pressure is less than the convergence tolerance. After the model converges, the pressure, flow rate and hydraulic impedance values ​​of the pipe section without installed sensors are extracted as inversion state parameters. The standardized deviation of the inverted hydraulic impedance from the normal reference is calculated using the following formula: ； in, Indicates pipe section The standardized deviation, This represents the pipe segment obtained from the inversion of the lightweight digital twin model. The average hydraulic impedance, Indicates pipe section The average hydraulic impedance under normal historical operating conditions It represents the standard deviation of hydraulic impedance under normal historical operating conditions.

6. The underground pipeline monitoring method as described in claim 5, characterized in that: The steps are as follows: Based on the spatiotemporal gradient anomaly index, the attribution information of the effective leakage event clusters, and the pipeline state parameters obtained by inversion from the lightweight digital twin model, the leakage probability of each pipeline segment is comprehensively evaluated, and the final leakage diagnosis result is output. For each pipe segment, the normalized maximum spatiotemporal gradient anomaly index, the effective leakage event cluster attribution indicator, and the standardized deviation of the inverted hydraulic impedance are calculated respectively. The three indicators are linearly weighted according to preset weights to obtain a comprehensive leakage probability score, expressed as: ; in, Indicates pipe section The comprehensive leakage probability score, Represents the normalized maximum spatiotemporal gradient anomaly index. For effective leak event cluster attribution indication, if the pipe section Belongs to any valid leakage event cluster, , , These are non-negative weighting coefficients; When the comprehensive leakage probability score exceeds the second threshold, it is determined that there is a leak in the pipe section; The output includes pipe segment identification, comprehensive leakage probability score, recommended inspection priority, and structured diagnostic results of the suspected leak initiation time.

7. The underground pipeline monitoring method as described in claim 6, characterized in that: The average hydraulic impedance under normal historical operating conditions The standard deviation, the standard deviation of the spatiotemporal gradient anomaly index, the first threshold and the second threshold are all dynamically updated based on the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

8. An underground pipeline network monitoring system, based on the underground pipeline network monitoring method according to any one of claims 1 to 7, characterized in that: include: The module includes a hydraulic impedance field construction module, a spatiotemporal gradient anomaly detection module, a water hammer wave event verification module, a digital twin state inversion module, a comprehensive leakage assessment module, and an adaptive threshold update module. The hydraulic impedance field construction module is used to calculate the hydraulic impedance of each pipe section based on pressure time series data and flow time series data, and to construct a hydraulic impedance spatiotemporal field covering the pipe network topology and time dimension. The spatiotemporal gradient anomaly detection module is used to extract spatial and temporal variation trends based on the spatiotemporal field of hydraulic impedance, fuse them to generate a spatiotemporal gradient anomaly index, and identify suspected leaking pipe sections based on the index to form a primary leak alarm event. The water hammer wave event verification module is used to determine the water hammer wave propagation characteristics by combining the topology of the underground pipe network and the physical properties of the pipe segments, to verify the consistency of the water hammer wave propagation law of multiple primary leakage alarm events, and to screen out the effective leakage event clusters that conform to the physical propagation logic. The digital twin state inversion module is used to drive the lightweight digital twin model to run by using pressure time series data, flow time series data and effective leakage event clusters as boundary conditions, and invert the pipeline network state parameters in areas where no sensors are deployed. The comprehensive leakage assessment module is used to calculate the comprehensive leakage probability score of each pipe segment based on the spatiotemporal gradient anomaly index, the attribution information of effective leakage event clusters, and the pipeline status parameters obtained by inversion from the digital twin model, and output the final leakage diagnosis result. The adaptive threshold update module is used to dynamically update the mean and standard deviation of hydraulic impedance, the standard deviation of spatiotemporal gradient anomaly index, the first threshold and the second threshold under historical normal operating conditions based on the current operating season of the pipeline network, the daily water load pattern or the service life of the equipment.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the file encryption method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the file encryption method according to any one of claims 1 to 7.

Images